This paper describes a content-based image retrieval digital library that s
upports geographical image retrieval over a testbed of 800 aerial photograp
hs, each 25 megabytes in size. In addition, this paper also introduces a me
thodology to evaluate the performance of the algorithms in the prototype sy
stem. The major contributions of this paper are two. 1) We suggest an appro
ach that incorporates various image processing techniques including Gabor f
ilters, image enhancement, and image compression, as well as information an
alysis technique such as self-organizing map (SOM) into an effective large-
scale geographical image retrieval system. 2) We present two experiments th
at evaluate the performance of the Gabor-filter-extracted features along wi
th the corresponding similarity measure against that of human perception, a
ddressing the lack of studies in assessing the consistency between an image
representation algorithm or an image categorization method and human menta
l model.